Introduction to model-based fault diagnosis and fault-tolerant control

Abstract

The automation of a process consists in providing a quasi-optimal solution to obtain the best possible quality of the final product and consequently an increase in profits. Automated system control theory has been widely developed and applied to industrial processes. These techniques ensure the stability of the closed-loop system and yield a pre-defined performance in the case where all system components operate safely. However, the more the process is automated, the more it is subject to the occurrence of faults. Consequently, a conventional feedback control design may result in an unsatisfactory performance in the event of malfunctions in the actuators, sensors, or other components of the system. This may even lead the system to instability. In highly automated industrial systems where maintenance or repair cannot always be achieved immediately, it is convenient to design control methods capable of ensuring nominal performance when taking into account the occurrence of faults. This control is referred to as fault-tolerant control (FTC) which has become of paramount importance in the last few decades. The design of an FTC system requires obviously quick fault detection and isolation (FDI) for adequate decision making. Hence, to preserve the safety of operators and the reliability of processes, the presence of faults must be taken into account during the system control design.